Notes on building the AI-native web.
Page 15 of 17
AIEO for a business that serves one area
“Near me” became “who should I use in…”. The answer is assembled from sources most local businesses never check.
Using an AI employee for research and briefs
One of the highest-value internal jobs, and the one where a confident fabrication does the most quiet damage.
Accessibility is a conversion feature
Argued as compliance it gets deferred forever. Argued as the reason people can complete your form, it gets done.
Internal linking is the lever nobody pulls
It costs nothing, requires no outreach, and it is how a pile of articles becomes a library that reads as authoritative.
Your documentation is marketing now
Buyers ask assistants how a product works before they contact you. The answer comes from your docs, if your docs are public.
Chatbot, copilot, agent: what the words actually mean
Vendors use them interchangeably, which makes buying hard. The distinctions that change what you get and what it costs.
Training an AI employee: what to actually feed it
Not your entire drive. The specific material that turns a generic assistant into something that sounds like your business.
The AI feature that should have been a form
Conversation is a slow interface for structured input. Sometimes the sophisticated answer is four fields and a button.
Should this be an agent or a workflow?
Agents are the exciting answer and usually the wrong one. The distinction that decides it is how much of the path you already know.
The AIEO mistakes that cost the most
Six errors, ordered by damage. Most are things being done deliberately, for reasons that made sense under the old rules.
What you should receive at the end of a build
The working software is the smallest part of it. Seven things to have agreed before the final invoice.
The entity graph starts with naming things consistently
Your company, your products, your categories. Every variant you use is a chance for a model to conclude you are two different things.